During the development of a child, most brain cells are generated before birth. There is one group of brain cells, however, that are generated in large numbers in infancy. This occurs in a brain region called the hippocampus. These cells are important for the normal development of memory, cognition and language, and disruption of these cells is present in children with autism. Disruption of these cells may also contribute to the prevalence of epilepsy in children with autism. Although it makes sense that these cells are responsible for key features of autism, children with the disease exhibit changes in many other parts of the brain as well. Our limited understanding of which changes are important, unfortunately, has slowed progress in the field. In the present proposal, therefore, we will examine the impact of eliminating a gene known to be involved in autism on these late-generated cells. These studies will reveal whether selectively disrupting these cells can reproduce features of autism. Demonstrating that these cells play an important role in the development of autism (or associated conditions like epilepsy), will provide a compelling rationale to develop new therapies to protect these vulnerable cells. In addition, since these cells are born so late in development, and are actually produced through childhood and into adulthood, it may be possible to prevent the disruption of cells born after a diagnosis of autism and thus allow the brain to naturally restore itself. It is our hope that these studies will move us closer to developing new and more effective treatments for autism. PUBLIC HEALTH RELEVANCE: The proposed studies seek to better elucidate the causes of autism by determining whether selective disruption of late-generated neurons - specifically hippocampal dentate granule cells - contributes to key features of autism. This will be achieved using a novel combination of transgenic mice to selectively eliminate a gene implicated in autism from these late-generated neurons. Significantly, although the proximal cause of autism remains to be determined in most cases, the late generation of these neurons (in the late embryonic period and in infancy) may make them uniquely vulnerable to a variety of insults faced by the newborn child (e.g. infection). The proposal will also examine the role of granule cells in the development of epilepsy, the condition most commonly associated with autism.